1 P2 partition

# Data partition used to fit the models
part <- "P2"

# Path
path= paste0("outputs/models/",part,"/")

1.1 Some posterior predictive checks

1.1.1 PPC Density overlay (on the log scale)

Comparison of the empirical distribution of the data \(y\) (dark line on the graphs) to the distributions of simulated/replicated data yrep from the posterior predictive distribution.

1.1.1.1 Predictions on the train dataset

Train dataset of the P2 partition => 28 provenances

1.1.1.2 Predictions of the test dataset

Test dataset of the P2 partition => 6 provenances

1.1.2 PPC violin plot

1.1.2.1 Predictions in each site

## $`1`

## 
## $`2`

## 
## $`3`

## 
## attr(,"class")
## [1] "list"      "ggarrange"

1.1.2.2 Predictions in each provenance

## $`1`

## 
## $`2`

## 
## $`3`

## 
## attr(,"class")
## [1] "list"      "ggarrange"

1.2 Residuals plots

1.2.1 In the train dataset

Predictions against predictive errors

1.2.2 In the test data set

Predictions against predictive errors